DE
DeepSeek
Language model · China ·Sep 2024

DeepSeek-V2.5

Language Open (Restricted)
236B
Parameters
7
Benchmarks

About DeepSeek-V2.5

DeepSeek-V2.5 is an AI model developed by DeepSeek, in the language category, released in 2024, made available as an open-weights model with 236B params.

On this page you'll find DeepSeek-V2.5's full specifications, including 7 benchmark results. Review provider pricing and benchmark scores below, or compare DeepSeek-V2.5 head-to-head with other language models.

Links

Run DeepSeek-V2.5 locally — what to buy

Share this build ↗

Sized to what DeepSeek-V2.5 actually needs (~8k context at Q4), not to the biggest GPU: Good = cheapest that runs it, Better = best value, Best = most headroom. Speed is a hardware estimate; anything past ~120 tok/s is shown as instant.

Software support

✓ measured · · compatible · — not supported. Informational only — speed is hardware-based.

Community

Would you run DeepSeek-V2.5 again?
No votes yet — be the first.
Did it run? — community reports by hardware

No reports yet. Own this model on some hardware? Be the first to confirm it runs.

Discussion

No comments yet. Start the discussion.

DeepSeek-V2.5 benchmark scores
Benchmark Category Score Variant
HumanEval code 89% python-from-hf-readme
LiveCodeBench code 41.8% jan-sep-2024-from-hf-readme
SWE-bench Verified code 16.8% from-hf-readme
Aider Benchmark coding 72.2 from-hf-readme
AlignBench v1.1 general 8.04 from-hf-readme
AlpacaEval 2 general 50.5% from-hf-readme
ArenaHard general 76.2% from-hf-readme

More models like DeepSeek-V2.5

Frequently asked questions

What is DeepSeek-V2.5?

DeepSeek-V2.5 is an AI model developed by DeepSeek, in the language category, released in 2024. It is tracked on GenAIList with its specifications, benchmark scores and provider pricing.

Who created DeepSeek-V2.5?

DeepSeek-V2.5 was developed by DeepSeek and released in 2024.

Is DeepSeek-V2.5 open source or proprietary?

DeepSeek-V2.5 ships as an open-weights model: you can download and self-host the weights, though the licence may place some restrictions on use.

How much does DeepSeek-V2.5 cost?

Pricing for DeepSeek-V2.5 depends on the provider. See the providers table on this page for the latest API rates.

How does DeepSeek-V2.5 perform on benchmarks?

DeepSeek-V2.5 is benchmarked across 7 evaluations on GenAIList, including HumanEval (89%). See the full benchmark table below and compare it with other models.

No videos yet. Know a good one? Add it below.

Know a great video about DeepSeek-V2.5?